Google MCP Server

What is Google MCP Server?

Google MCP Server is an MCP (Model Context Protocol) server that integrates Google Custom Search API with AI assistants and other MCP‑compatible clients, enabling web and image search through standardized tool calls.

How to use Google MCP Server?

Clone the repository, install dependencies (npm install), and set up a .env file with your GOOGLE_API_KEY and GOOGLE_CX_ID. Then run a test search with npx ts-node src/test.ts or add the server to your MCP client’s configuration (e.g., claude_desktop_config.json) specifying the command, arguments, and environment variables.

Key features of Google MCP Server

• Provides web search via googleSearchContent tool.
• Provides image search via googleSearchImages tool.
• Returns structured results (titles, links, snippets).
• Returns Markdown image link for first image result.
• Works with any MCP‑compatible client.

Use cases of Google MCP Server

• Letting an AI assistant perform real‑time web searches.
• Enabling an AI to retrieve image links based on a query.
• Integrating Google search capabilities into custom MCP workflows.
• Testing search functionality before adding the server to a client.

FAQ from Google MCP Server

What runtime is required?

Node.js 18 or newer and npm are required.

How do I obtain the necessary credentials?

You need a Google Custom Search API key and a Custom Search Engine ID. Obtain them from the Google Custom Search JSON API documentation.

Where are the API credentials stored?

In a .env file in the project root (GOOGLE_API_KEY and GOOGLE_CX_ID) or passed as environment variables in the MCP client configuration.

What transports does this server use?

The README does not specify transport details; it is assumed to use standard MCP transport (e.g., stdio) as typical for MCP servers invoked via npx.

Can I modify the code?

Yes. Modify files in the src directory, update dependencies in package.json if needed, and restart your MCP client or test script to apply changes.